OpenMS
2.4.0
|
Configuration class for MSNumpress. More...
#include <OpenMS/FORMAT/MSNumpressCoder.h>
Public Member Functions | |
NumpressConfig () | |
desired mass accuracy for linear encoding (-1 no effect, use 0.0001 for 0.2 ppm accuracy @ 500 m/z) More... | |
void | setCompression (const std::string &compression) |
set compression using a string mapping to enum NumpressCompression. More... | |
Public Attributes | |
double | numpressFixedPoint |
double | numpressErrorTolerance |
fixed point for numpress algorithms More... | |
NumpressCompression | np_compression |
check error tolerance after encoding, guarantee abs(1.0-(encoded/decoded)) <= this, 0=do not guarantee anything More... | |
bool | estimate_fixed_point |
which compression schema to use More... | |
double | linear_fp_mass_acc |
whether to estimate the fixed point or use the one proved with numpressFixedPoint More... | |
Configuration class for MSNumpress.
Contains configuration options for ms numpress
|
inline |
desired mass accuracy for linear encoding (-1 no effect, use 0.0001 for 0.2 ppm accuracy @ 500 m/z)
|
inline |
set compression using a string mapping to enum NumpressCompression.
compression | A string from NamesOfNumpressCompression[] |
Exception::InvalidParameter | if compression is unknown. |
References seqan::find().
bool estimate_fixed_point |
which compression schema to use
double linear_fp_mass_acc |
whether to estimate the fixed point or use the one proved with numpressFixedPoint
NumpressCompression np_compression |
check error tolerance after encoding, guarantee abs(1.0-(encoded/decoded)) <= this, 0=do not guarantee anything
double numpressErrorTolerance |
fixed point for numpress algorithms
double numpressFixedPoint |